SMA crossover with ADX filter passes all 6 gates

We tested a classic SMA golden cross (21 over 50) with one twist: it only trades when ADX stays below 55, filtering out strong trends. The strategy passed all 6 gates with a 2.2% average profit per trade, a 100% win rate in both training and validation, and a deflated Sharpe ratio of 11.55. The fluke probability after multi-test correction landed at 0.2%, well under our 50% threshold.

The logic is simple. We enter long when the 21-period SMA crosses above the 50-period SMA, but only if ADX signals low conviction, meaning the market is ranging or consolidating rather than trending hard. Exits happen when net profit hits the target. No short trades. The backtest ran on 15-minute bars across BTC, ETH, SOL, XRP, DOGE, BNB, ADA, AVAX, LINK, and LTC.

Here is where you should stay skeptical. The validation phase logged zero forward test trades, so we have no live confirmation yet. A 100% win rate in backtesting with a 2.2% average gain is unusually clean, and that can mean the strategy fired rarely or caught a specific regime feature that might not repeat. The deflated Sharpe ratio accounts for multiple testing, but we still recommend watching how it performs once live trades start coming in.

The ADX filter is doing heavy lifting here. ADX below 55 typically corresponds to sideways or weakly directional markets, where mean reversion setups like crossovers tend to work better than in explosive trends. If we enter a prolonged high-ADX environment across all ten coins, this strategy will sit idle. That is by design, but it also means coverage is conditional.

We added this to the survivor library because it cleared every statistical gate we throw at candidates, including the toughest one: deflated Sharpe after correcting for the hundreds of other strategies tested in the same phase. The math says it is probably not a fluke. The lack of forward trades says we need more time to confirm.

Check the full survivor registry at stratproof.com/survivors to see how this strategy compares to others that graduated, or head to stratproof.com/prove to test your own ADX-filtered crossover variant with different thresholds.